Attenuation correction has recently been proposed in order to increase sensitivity and specificity for Tc-99m MIBI perfusion scans. However, the initial publications validating this increase in sens/spec with attenuation correction used a specially designed system that avoided the possibility of scattered Tc photons contaminating the attenuation data. The commercially implemented versions of this system do not avoid this possibility of contamination. We sought to measure the effect of this contamination on clinical cardiac studies, in order to determine how accurate a correction was necessary to achieve a clinically accurate perfusion image. Preliminary results for this project were accepted for oral presentation at the Society of Nuclear Medicine meeting in June, 1998. A full manuscript is being prepared.